Infinity, NaN, and divide function

[RMDNA]

All,

In a custom column, I'm attempting to divide two columns, [Committed] and [Actual] (fixed decimal numbers in the backend, as U.S. Currency), and attempting to get a [% to target] result. The end goal is to filter by these categories:

a. commited for $0, spent $0

b. commited for $0, spent >$0

c. committed >$0, spent $0

d. commited for $x, spent $y

Trying to divide, either by [A]/[B] or DIVIDE( ), I'm getting a mix of actual zeroes, NaNs, and infinities, and I haven't been able to use conditional columns or if/then statements to group them. Any help?

[v-qiuyu-msft]

Hi @RMDNA,

In my opinion, it should be three categories, as both a. and c. have the numerator 0 which can be treated as the same scenario. Based on my tested, use DIVIDE() function, it returns three kind of values, blank, zero and actual decimal.

If the above is not your required results, please share some sample data and screenshots about desired results for our analysis.

Best Regards,
Qiuyun Yu

[akaratr]

Thanks. I also have the same issue. It's solved.

[Greg_Deckler]

Can you post some sample data? Generally, if zeros are involved, use DIVIDE to avoid divide by zero errors. Additionally, use the erd parameter "alternateresult" option of the DIVIDE function to catch divide by zero and return what you want. Potentially wrap it all in an IF statement if necessary.